these blog posts are written by Dr. alfredo G. Tomasselli, a Retired bio-Chemist and Bio-physicist who spent his life in Academia and The Pharmaceutical Industry researching cures and treatments for diseases such as HIV/Aids, Diabetes, and Rheumatoid arthritis. His daughter sara inspires this blog and helps to write posts. sara was diagnosed with type 1 diabetes on april 4th, 2014 at the age of 30. 

Dec 16 2016 Towards new therapies for Type 1 Diabetes: Scientists found approved medical drugs able to convert pancreatic α-cells into β-like-cells that produce insulin

Towards new therapies for Type 1 Diabetes: Scientists found approved medical drugs able to convert pancreatic α-cells into β-like-cells that produce insulin

This document is for informational purposes only and can not be used to make medical or other personal decisions.

Type 1 diabetes is an autoimmune disease that occurs when certain cells of the immune system go awry and mistakenly recognize and kill (or inactivate) the insulin-producing β-cells. The β-cells are housed within the islets of Langheran along with other cell types such as the glucagon producing α-cells; other cell types are also present (footnote i). β-cells depletion leads to the physiologically insufficient production of insulin and persistent high blood glucose. The disease is manageable with daily insulin injections and a healthy lifestyle, but not curable. A cure would require the regeneration of the insulin-producing β-cells and the reeducation of the patient’s faulty immune system and significant progress is being made. Two research groups, one in Austria led by Dr. Stefan Kubicek (Li et al, 2016, Reference 1) and the other in France led by Dr. Patrick Collombat (Ben-Othman et al., 2016, Reference 2) have succeeded in converting α-cells into β-like cells (α-to-β-like cells) in animal models of diabetes and the newly generated β-like cells were able to restore normal blood glucose in these animals. Moreover, the conversion process was also achieved in purified human and animal islets. Importantly, in both studies, it was observed that the α-cells converted to β-like cells were naturally replenished. To accomplish the conversion of α-to-β-like cells, the former group used two approved anti-malaria medical drugs called artemisinins (artemether, and dihydroartemisinin) while the latter employed the food supplement gamma-amino butyric acid (GABA). Other important points of both studies were the identification of key molecular players and their putative role in driving the process.

According to the first study artemether binds to the protein gephyrin which, in turn, interacts with and activates the GABA receptor signaling in α-cells. This triggers the ejection, from the α- cell nucleus, of the transcription factor ARX causing its inactivation. This series of events leads to the α-to-β-like cells conversion consisting in the inhibition of glucagon release from α-cells and in the increase of their insulin production. The authors of the second study observed that treatment of the animals, or islets, with gamma-amino butyric acid (GABA), was also likely to lead to the activation of its GABA receptor. This activation resulted in the downregulation (reducing the amount) of the transcription factor ARX and consequently the conversion of α-to- β-like cells, similar to the first study.

These studies are interesting, but are still at the animal model stage; only clinical studies with Type 1 Diabetes people may tell us if any of these molecules work.

The main feature of Type 1 Diabetes (T1D) is the inability of the body to produce insulin, a protein hormone that regulates glucose levels. Insulin is produced by the β-cells housed within the islets of Langheran (footnote i) of the pancreas. However, in people with T1D the β-cells are mistakenly recognized and killed, or disabled, by certain white blood cells (Leukocytes, footnote ii) of an immune system that has gone awry. Without the production of physiologically sufficient insulin the body’s glucose raises to life-threatening levels. Daily injections of insulin and a strict lifestyle are the standard treatment of T1D, but it is difficult and annoying to closely adhere to them; and a poor compliance brings the risk of serious health complications. At the present, the only ways to stay off insulin are a total pancreas transplant (a very demanding surgery), or islets transplant (a relatively mild procedure), but these clinical procedures are limited to a few thousand patients because of a shortage of pancreas donors and both transplants have numerous shortcomings (described in Type 1 Diabetes ResearchAdvances To Develop A Cure, part 3).

In previous postings we have talked about successes and drawbacks in the development of technologies to prepare large quantities of β-cells from stem cells, that can be safely transplanted to many patients; we have also addressed strategies to reeducate the faulty immune system that leads to T1D and to bypass the autoimmune process caused by the introduction of foreign islets into the host; (described in Type 1 Diabetes Research Advances To Develop A Cure, parts, 4, 7, and 6, respectively).

In this posting we report about two recently published studies regarding the conversion of pancreatic α-cells into β-like cells (α-to-β-like cells) accomplished in animal models and in islets purified from people and animals. Both studies succeeded in carrying out this conversion by using approved medical drugs and in gaining useful information regarding the series of molecular events guiding the process. One study was led by Dr. Stefan Kubicek at CeMM Research Center for Molecular Medicine of the Austrian Academy of Sciences, Vienna (Austria) (Li et al, 2016, Reference 1) and the other by Dr. Patrick Collombat Universite´ Coˆte d’Azur, CNRS, Inserm, Nice (France) (Ben-Othman et al., 2016, Reference 2), but it looks like parts of the studies were collaborations between the two groups.

The two research groups capitalized on the large amount of knowledge obtained from previous studies regarding the molecules and biological events involved in the differentiation of embryonic stem cells and their maturation into cells that form the pancreas. In the intricate network of molecular players governing the pancreas formation, it was found that two transcription factors (see footnote iv), PAX 4 and ARX, foster the formation of β- and α- cells, respectively, from precursor cells. Moreover, it was shown that the α-to-β-like cells conversion could be driven by misexpressing (see footnote v) PAX 4 in α-cells or by downregulating (reducing or suppressing) ARX expression in α-cells. Due to the findings that downregulation of ARX expression was the main trigger of the α-to-β-like cells conversion, they focused on ARX.

The study by the Kubicek group (Li et al., 2016, Reference 1) describes a cellular screening in a cell line engineered with the purpose to identify compounds that would repress ARX gene expression. The compounds repressing ARX gene expression were then tested in an α-cell line (referred as to αTC1 cells) to see if they would make these cells produce insulin. They tested 280 approved medical drugs in the screen and found two artemisinins (anti-malaria drugs), named artemether, and its active metabolite, named dihydroartemisinin, to fully inhibit the ARX gene overexpression and were able, as expected, to induce the α-cells to produce insulin, they accomplished the α-to-β-like cell conversion. They also identified major cellular players involved in the conversion and how they interact with each other which allowed them to suggest a possible mechanism by which the conversion process took place: artemether binds to the protein gephyrin and stabilizes it; gephyrin-bound artemether interacts and activates GABA receptor signaling in α-cells; active GABA receptor signaling drives the expulsion of the transcription factor ARX from the nucleus to the cytoplasm which results in ARX inactivation. This series of events leads to the α-to-β-like cells conversion consisting in the inhibition of glucagon release from α-cells and in the increase of their insulin production.

Then they tested artemether for its ability to perform α-to-β-like cells conversion in Zebrafish and rodent animal models whose real β-cells were depleted in the laboratory. After a 4-day treatment of the β-cell depleted Zebrafish they found that the β cells of this animal model increased by about 75% compared to non treated animals. Also, glucose levels in β-cell depleted Zebrafish increased by 34% over normal Zebrafish. However, upon artemether treatment Zebrafish had 42% lower glucose levels compared to normal animals. Moreover, in β-cell depleted rats a 23 days artemether treatment also resulted in a drastic reduction of blood glucose levels compared to non-treated animals.

Having shown success in animal models, these scientists focused on intact human islets (apparently from donors without T1D). Artemether treatment of human islets in vitro (in Petri dishes) resulted in the reduced expression of α-cell transcription factor ARX and increased insulin content compared to the untreated islets from the same donor. Moreover, in the presence of high glucose concentrations, significant more insulin was secreted by artemether- treated islets than from those non-treated.

Collectively, this study shows that the anti-malaria drug artemether generates pancreatic β-like cells from α-cells by the putative mechanism described above where activation of the GABA receptor signaling is a critical feature.

The group led by Dr. Collombat (Ben-Othman et al. 2016, Reference 2) first focused on the treatment of wild type mice (non-diabetic mice) with amino-butyric acid (GABA). Even at a relatively low GABA concentration (0.250 mg GABA/kg mice, by intraperitoneal injections), they observed, compared to non-treated animals (controls), doubling of the number of islets, an increase in the numbers of glucagon cells, and most importantly, a massive increase in the number of cells containing insulin (from here on referred as to insulin+ cells). The process was characterized by a decrease in ARX transcript concentration, and these scientists emphasized the critical importance of the downregulation of ARX in the α-to-β-like cells conversion. When the GABA animals were challenged by high glucose amounts, they responded more effectively than controls in lowering their glucose levels, an indication that the increased insulin+ cells were functionally β-like cells. They suggested that the downregulation (reduction) of the transcription factor ARX upon treatment with gamma-amino butyric acid (GABA) is most likely caused by the activation of the GABA receptor, a conclusion similar to that achieved by the last group.

Then they turned to mice that had been rendered diabetics because they had lost most of their β cells upon chemical treatment with streptozotocin. These animals were allowed to reach a concentration of blood glucose ~300mg/dL (hyperglycemia) and then treated with solutions containing either GABA or salt (control). The animals that received GABA regenerated their β-like cells over 70-95 days, while the controls did not. Moreover, the GABA animals responded to high doses of glucose and returned to normal glucose levels. They emphasized that they were able to go through two cycles of β-cells killing and GABA treatment and were able to regenerate them in every cycle. It is important to note that the α-cells lost to generate β-like cells were continually replaced with fresh α-cells.

Then they wanted to find out whether or not the α-to-β-like cells conversion would occur in human islets. Indeed, they observed that culturing human islets for 14 days in the presence of GABA resulted in a 37% decrease in α-cells and a 24% increase in β-cells indicating conversion. Next, they took 500 human islet equivalents and transplanted them under the kidney capsule of immunodeficient mice. Upon 1 month treatment with GABA, or saline, they observed a decrease in α-cells and an increase in β-cells, another indication of a successful α-to-β-like conversion.

In conclusion, these studies are interesting, but big questions still remain on whether the α-to- β-like cells conversion fostered by gamma-amino butyric acid (GABA) and artemisinins is going to translate from animal models to humans, and if so, how efficiently; how much of these drugs are required and for how long do they need to be used to revert the disease? How close are the β-like cells to the lost natural β-cells? Hopefully, they will be close enough to functionally replace them efficiently, but sufficiently different to avoid the autoimmune attack that starts T1D. If an autoimmune attack would occur, hopefully it would be mild and possibly the lost β-like cells will be continuously replaced by newly formed ones.

Only clinical trials with T1D people, hopefully starting soon, will tell how close we are to realize the dream of normalizing blood sugar by regenerating β-cells directly in the patient by using simple molecules. Let’s hope it does not take long to know the answers.


i) The islets of the Langheran are small (about 0.2 mm diameter) clusters of cells disseminated across the pancreas constituting 1-2% of its mass. They contain glycemic regulatory hormones producing cell species such as: (I) β-cells that produce insulin (each islet contains approximately 1,000 β-cells, i.e., they constitute the majority of the cells of the islet); and (II) α-cells that produce glucagon that raises the glucose concentration in the blood, thus opposing the insulin glucose-lowering activity; also present are the three cell types, 𝛿-, ε-, and PP-cells, each one of them secretes specific hormones.

ii) The CD8+ T leukocytes are the major immune cells that infiltrate the islets of Langheran, and are regarded as the most prominent killers of the insulin-producing β-cells; yet, CD4+ leukocytes, B leukocytes and macrophages are also present in the infiltrate and take part in β-cells elimination. Collectively, the infiltration by these leukocytes is referred to as insulitis and it is limited to the islets of Langheran. Typically, the actions of the above T cells are kept in check by another subtype of T cells called regulatory T cells (or Treg). It has been shown that in autoimmune diseases, such as T1D, Treg cells are defective.

iii) Autoimmunity is an immune response whereby a person’s immune system attacks the body’s own cells causing their destruction, e.g., in T1D the β-cells are destoyed; Alloimmunity is an immune response against a foreign entity (non-self), e.g., a transplant from one person to another of an organ like the pancreas, or of pancreas’ islets.

iv) A transcription factor is a protein that binds to a specific DNA sequence to control the rate of its transcription into messenger RNA which consequently may be translated into the generation of a protein.

v) It is called misexpression because it refers to a gene artificially expressed in cells (here it is α- cells) where it does not naturally belong.


Ben-Othman, N., Vieira, A., Courtney, M., Record, F., Gjerns, E., Avolio, F., Hadzic, B., Druelle, N., Napolitano, T., Navarro-Sanz, S., et al. (2016). Longterm GABA administration induces alpha cell- mediated beta-like cell neogenesis. Cell 168. Published online December 1, 2016.  10.1016/j.cell.2016.11.002.

Li, J., Casteels, T., Frogne, T., Ingvorsen, C., Honore´ , C., Courtney, M., Huber, K.V.M., Schmitner, N., Kimmel, R.A., Romanov, R.A., et al. (2016). Artemisinins target GABAA receptor signaling and impair alpha cell identity. Cell 168. Published online December 1, 2016.

2017 Progress on reprogramming liver cells into pancreas β-like cells

2016 Study Reports that Sweetened Beverages are Associated with an Increased Risk of both Latent Autoimmune Diabetes in Adults and Type 2 Diabetes